A global provider of financial software, data, and workflow platforms was preparing to move IBM MQ workloads off aging UK hardware and onto newer virtualized infrastructure. Operationally, that was the right move. Commercially, it was dangerous. IBM MQ server licensing is measured in Processor Value Units (PVUs), and under full-capacity rules a move to higher-core hosts can make the apparent license requirement grow far faster than the actual application workload.
That risk became real as the client reviewed its IBM MQ position across UK and US entities. Depending on which host assumptions and agreement scope were used, the numbers moved dramatically. An early UK-only full-capacity model showed a 1,170-PVU gap and $176,904 of modeled exposure. A later draft ELP still showed a 350-PVU discrepancy because IBM’s review was working without ILMT and with incomplete host mapping on part of the environment.
The client brought in UMS for audit defense and licensing strategy support. UMS rebuilt the fact base, corrected the hardware assumptions driving the gap, and narrowed the modeled UK shortfall to 50 PVU with $7,560 of modeled exposure. Just as important, UMS helped the team remediate the remaining requirement with targeted IBM purchases instead of buying against the worst-case full-capacity scenario.
The Challenge
IBM MQ was not the only moving part. The licensing problem sat at the intersection of infrastructure change, agreement sprawl, and audit-style scrutiny:
- Legacy hardware was being replaced by larger virtual hosts — The new hosts had more physical cores than the older machines they replaced. If the client licensed the environment at full capacity without careful modeling, the MQ requirement could jump sharply even though the workloads themselves were still modest.
- Entitlements were split across multiple IBM agreement footprints — UK and US Passport Advantage records did not present as a single clean pool. That meant scope decisions mattered, and a conservative reading could make the estate look more under-licensed than it really was.
- Draft audit records were built on incomplete infrastructure assumptions — IBM’s draft ELP explicitly noted that ILMT was not deployed and that some physical-host details had to be assumed. In IBM licensing, those assumptions are expensive because they usually push the position toward full-capacity counting.
- The business still had to keep production messaging running — MQ supported business-critical workflows. The company could not pause the migration just to untangle licensing math, and it could not afford to overbuy licenses simply to make the problem go away.
The core issue was not just compliance. It was decision quality. The client needed to know which number represented a true licensing requirement, which number was only a conservative scenario, and which remediation steps were actually worth buying.
How UMS Solved It
UMS approached the engagement the same way it handles any high-stakes publisher review: isolate the facts, pressure-test the assumptions, and buy only what the validated position requires.
Step 1: Rebuild the deployment and entitlement picture
UMS pulled together host details, VM mappings, IBM order history, and Passport Advantage records to understand where MQ was deployed and which entitlements were already owned. That work mattered because one incorrect processor model or one assumed physical host could add hundreds of PVUs to the apparent requirement.
Step 2: Model the position under multiple licensing scenarios
Rather than accept a single audit number at face value, UMS compared the estate under the different paths the client was actually considering:
| Scenario | What it assumed | Result |
|---|---|---|
| Early UK full-cap model | 20 UK cores / 1,880 PVU deployment | 1,170-PVU gap and $176,904 modeled exposure |
| Refined UK full-cap model | Corrected host details / 760 PVU deployment | 50-PVU gap and $7,560 modeled exposure |
| Future global sub-cap model | ILMT-based virtual-core reporting | 2,110-PVU gap versus 7,770 under full capacity |
This was the turning point in the engagement. UMS showed that the biggest number on the table was not the inevitable outcome. It was the result of conservative infrastructure assumptions that could be tested and, in several cases, disproven.
Step 3: Convert the analysis into scoped remediation
Once the exposure was narrowed, UMS helped the client close the remaining gaps with precise purchases rather than a blanket license expansion. The documented commercial actions included:
- A March 30, 2022 IBM UK quote for 100 PVU of IBM MQ Software Subscription & Support reinstatement at GBP 5,208 before tax.
- A June 9, 2022 IBM US quote, followed by a June 30, 2022 invoice and order confirmation, for 280 PVU of IBM MQ licenses at $29,120 before tax.
Those records matter because they show the analysis did not stop at a spreadsheet. It translated into actual procurement actions sized to the validated requirement.
Step 4: Build the longer-term playbook
UMS also documented the path for keeping future MQ growth from drifting back into full-capacity overbuy. In later planning, UMS showed that a global full-cap model would create a 7,770-PVU gap at roughly $932,400 list price, while a sub-capacity approach with ILMT would reduce the modeled gap to 2,110 PVU at roughly $253,200 list price. That is a 73% lower modeled growth gap, plus a drop in projected annual support from about $227,787 to $91,211.
In other words, UMS did not just help close the immediate issue. It gave the team a repeatable rule set for the next round of virtualization, growth, and publisher scrutiny.
Results
| Metric | Before | After | Impact |
|---|---|---|---|
| Modeled UK IBM MQ licensing gap | 1,170 PVU | 50 PVU | 96% lower |
| Modeled UK exposure | $176,904 | $7,560 | $169,344 lower |
| UK remediation path | Broad full-cap purchase risk | 100-PVU reinstatement quote | Targeted closeout |
| US remediation path | No scoped purchase plan | 280 PVU purchased | Precisely matched requirement |
The practical outcome was clarity. The client no longer had to respond to IBM MQ licensing with a worst-case purchase strategy. UMS separated modeled exposure from validated requirement, then helped the company remediate the remaining need through specific UK and US transactions instead of licensing every possible physical core that might someday host MQ.
Key insight: In IBM MQ reviews, the most expensive mistake is often not the workload itself. It is letting incomplete host data and full-capacity assumptions define the commercial response.
Additional Outcomes
- Future virtualization got a safer playbook — UMS showed that ILMT-based sub-capacity reporting could reduce modeled future gap exposure from 7,770 PVU to 2,110 PVU across broader MQ deployment plans.
- Support-cost risk dropped along with license risk — The same sub-capacity model lowered projected annual support from about $227,787 to $91,211, which matters in multi-year budgeting.
- The team got a cleaner buying standard — UMS confirmed that MQ Client carried no license cost while MQ Server remained the fee-bearing component, preventing unnecessary spend on the wrong product category.
The same principle applies across publisher-led reviews: when licensing detail is validated early, modeled exposure does not have to become the commercial outcome.